International Meeting for Autism Research (London, May 15-17, 2008): Comparison of production of facial expressions by children with and without Autism

Comparison of production of facial expressions by children with and without Autism

Saturday, May 17, 2008
Champagne Terrace/Bordeaux (Novotel London West)
10:30 AM
A. Biswas , Psychology, University of Sheffield, Sheffield, United Kingdom
P. Mitchell , Psychology, University of Nottingham, United Kingdom
O. Pascalis , Psychology, University of Sheffield, United Kingdom
Background: Facial expression production represents a vital part of non-verbal communication in everyday life which improves during development (Herba and Phillips, 2004). When presented on pictures, happy expressions are recognised at the age of two, sad and anger at four to five whereas fear, surprise and disgust are recognised only around the age of 10 (Gross and Baliff, 1991).

One of the cardinal features of Autism is atypical face processing, meaning complete disinterest in faces (Klin et al., 1999) and lack of facial expressions (ICD 10 and ADI-R).  

Objectives: our aim was to compare facial expression production and understanding, in children with Autism (mean age of 12.1 and mean IQ of 96.3) and their matched controls.

Methods: Participants were filmed in the following situations: a-Baseline task - participants were asked to copy facial movements that are required to produce the 6 basic facial expressions to determine if the voluntary movement were possible. b-Participants were then asked to produce expressions on demand without any audio or visual cue. c-Lastly, twelve short stories were presented and participants were asked to label, then to produce each emotion felt by the main character.

Results: Two raters scored all videos. Chi-square tests showed that baseline tasks were performed well above chance level with no significant group differences. For production of facial expressions both with and without context and for emotion labelling tasks split-plot ANOVAs were performed. Both groups were equally good at producing happy, fear and anger but the autistic participants were significantly worse for disgust, surprise and sad. If labelling helps the overall production's performance in typical population it failed to do so in children with Autism with the exception of happiness.

Conclusions: Complex emotion production improved in typically developing children within appropriate context but failed to do so in children with Autism.